technology-selection

Guide .NET 8+ developers in selecting AI/ML technology stacks.

Updated Mar 19, 2026
One-click install
npx skills add https://github.com/kierunb/ReliableAppDemo --skill technology-selection-kierunb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: technology-selection
Source: https://github.com/kierunb/ReliableAppDemo/tree/main/.github/skills/dotnet-ai/skills/technology-selection
Command: npx skills add https://github.com/kierunb/ReliableAppDemo --skill technology-selection-kierunb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps development teams choose the most suitable AI/ML technology stack for .NET 8+ projects, aligning capabilities with project goals, team skills, and deployment constraints.

Core Features & Use Cases

  • Provides a decision-tree-based guide across ML.NET, Microsoft.Extensions.AI (MEAI), Microsoft Agent Framework, and GitHub Copilot SDK to cover traditional ML, chat-based orchestration, and tool-enabled workflows.
  • Recommends a layered architecture that sequences abstraction (MEAI) → provider SDKs → orchestration, with guardrails and best practices for security, cost management, and deployment considerations.
  • Use case: A web application needs both a structured model for numerical predictions and an LLM-driven assistant; the guide helps select the stack and configure DI-based services, retries, and token budgeting.

Quick Start

Assess a .NET 8+ project and output a recommended AI/ML technology stack with rationale.

Frequently Asked Questions about technology-selection

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the right .NET AI stack for my application?

To choose the right .NET AI stack, use a decision-tree guide that evaluates ML.NET for structured data, MEAI for chat orchestration, and Microsoft Agent Framework for complex workflows based on your specific project goals and constraints.

When should I use ML.NET vs Microsoft.Extensions.AI for a .NET project?

Use ML.NET for structured data and numerical predictions, whereas Microsoft.Extensions.AI (MEAI) is designed for chat-based orchestration and tool-enabled workflows in .NET 8+ applications.

How do I configure dependency injection and token budgeting for .NET AI services?

Configure dependency injection and token budgeting by implementing a layered architecture that sequences MEAI abstractions first, followed by provider SDKs and orchestration frameworks with built-in guardrails.

Does ONNX Runtime work with OllamaSharp for local AI inference in .NET?

ONNX Runtime and OllamaSharp are both supported local inference options for .NET AI projects, allowing you to run models on-premises while integrating with cloud-based provider SDKs when needed.

Can I use GitHub Copilot SDK alongside Microsoft Agent Framework in web apps?

Yes, the GitHub Copilot SDK handles Copilot integrations while Microsoft Agent Framework manages complex workflows, and both can be configured together across web and service apps using MEAI abstractions.

What are the limitations of using MEAI without a provider SDK in .NET?

Using MEAI without a provider SDK limits direct model access and orchestration capabilities, as the recommended layered architecture sequences MEAI abstractions before provider SDKs to properly enforce guardrails and manage deployment costs.